Development of an optical character recognition pipeline for handwritten form fields from an electronic health record
Identifieur interne : 000380 ( Main/Merge ); précédent : 000379; suivant : 000381Development of an optical character recognition pipeline for handwritten form fields from an electronic health record
Auteurs : Luke V. Rasmussen [États-Unis] ; Peggy L. Peissig [États-Unis] ; Catherine A. Mccarty [États-Unis] ; Justin Starren [États-Unis]Source :
- Journal of the American Medical Informatics Association : JAMIA [ 1067-5027 ] ; 2011.
Abstract
Although the penetration of electronic health records is increasing rapidly, much of the historical medical record is only available in handwritten notes and forms, which require labor-intensive, human chart abstraction for some clinical research. The few previous studies on automated extraction of data from these handwritten notes have focused on monolithic, custom-developed recognition systems or third-party systems that require proprietary forms.
We present an optical character recognition processing pipeline, which leverages the capabilities of existing third-party optical character recognition engines, and provides the flexibility offered by a modular custom-developed system. The system was configured and run on a selected set of form fields extracted from a corpus of handwritten ophthalmology forms.
The processing pipeline allowed multiple configurations to be run, with the optimal configuration consisting of the Nuance and LEADTOOLS engines running in parallel with a positive predictive value of 94.6% and a sensitivity of 13.5%.
While limitations exist, preliminary experience from this project yielded insights on the generalizability and applicability of integrating multiple, inexpensive general-purpose third-party optical character recognition engines in a modular pipeline.
Url:
DOI: 10.1136/amiajnl-2011-000182
PubMed: 21890871
PubMed Central: 3392858
Links toward previous steps (curation, corpus...)
- to stream Pmc, to step Corpus: 000121
- to stream Pmc, to step Curation: 000121
- to stream Pmc, to step Checkpoint: 000132
- to stream Ncbi, to step Merge: 000114
- to stream Ncbi, to step Curation: 000114
- to stream Ncbi, to step Checkpoint: 000114
Links to Exploration step
PMC:3392858Le document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en">Development of an optical character recognition pipeline for handwritten form fields from an electronic health record</title>
<author><name sortKey="Rasmussen, Luke V" sort="Rasmussen, Luke V" uniqKey="Rasmussen L" first="Luke V" last="Rasmussen">Luke V. Rasmussen</name>
<affiliation wicri:level="2"><nlm:aff id="aff1">Biomedical Informatics Research Center, Marshfield Clinic Research Foundation, Marshfield, Wisconsin, USA</nlm:aff>
<country xml:lang="fr">États-Unis</country>
<wicri:regionArea>Biomedical Informatics Research Center, Marshfield Clinic Research Foundation, Marshfield, Wisconsin</wicri:regionArea>
<placeName><region type="state">Wisconsin</region>
</placeName>
</affiliation>
</author>
<author><name sortKey="Peissig, Peggy L" sort="Peissig, Peggy L" uniqKey="Peissig P" first="Peggy L" last="Peissig">Peggy L. Peissig</name>
<affiliation wicri:level="2"><nlm:aff id="aff1">Biomedical Informatics Research Center, Marshfield Clinic Research Foundation, Marshfield, Wisconsin, USA</nlm:aff>
<country xml:lang="fr">États-Unis</country>
<wicri:regionArea>Biomedical Informatics Research Center, Marshfield Clinic Research Foundation, Marshfield, Wisconsin</wicri:regionArea>
<placeName><region type="state">Wisconsin</region>
</placeName>
</affiliation>
</author>
<author><name sortKey="Mccarty, Catherine A" sort="Mccarty, Catherine A" uniqKey="Mccarty C" first="Catherine A" last="Mccarty">Catherine A. Mccarty</name>
<affiliation wicri:level="2"><nlm:aff id="aff2">Essentia Institute of Rural Health, Duluth, Minnesota, USA</nlm:aff>
<country xml:lang="fr">États-Unis</country>
<wicri:regionArea>Essentia Institute of Rural Health, Duluth, Minnesota</wicri:regionArea>
<placeName><region type="state">Minnesota</region>
</placeName>
</affiliation>
</author>
<author><name sortKey="Starren, Justin" sort="Starren, Justin" uniqKey="Starren J" first="Justin" last="Starren">Justin Starren</name>
<affiliation wicri:level="2"><nlm:aff id="aff1">Biomedical Informatics Research Center, Marshfield Clinic Research Foundation, Marshfield, Wisconsin, USA</nlm:aff>
<country xml:lang="fr">États-Unis</country>
<wicri:regionArea>Biomedical Informatics Research Center, Marshfield Clinic Research Foundation, Marshfield, Wisconsin</wicri:regionArea>
<placeName><region type="state">Wisconsin</region>
</placeName>
</affiliation>
<affiliation wicri:level="2"><nlm:aff id="aff3">Division of Health and Biomedical Informatics, Department of Preventive Medicine, Northwestern University, Chicago, Illinois, USA</nlm:aff>
<country xml:lang="fr">États-Unis</country>
<wicri:regionArea>Division of Health and Biomedical Informatics, Department of Preventive Medicine, Northwestern University, Chicago, Illinois</wicri:regionArea>
<placeName><region type="state">Illinois</region>
</placeName>
</affiliation>
</author>
</titleStmt>
<publicationStmt><idno type="wicri:source">PMC</idno>
<idno type="pmid">21890871</idno>
<idno type="pmc">3392858</idno>
<idno type="url">http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3392858</idno>
<idno type="RBID">PMC:3392858</idno>
<idno type="doi">10.1136/amiajnl-2011-000182</idno>
<date when="2011">2011</date>
<idno type="wicri:Area/Pmc/Corpus">000121</idno>
<idno type="wicri:Area/Pmc/Curation">000121</idno>
<idno type="wicri:Area/Pmc/Checkpoint">000132</idno>
<idno type="wicri:Area/Ncbi/Merge">000114</idno>
<idno type="wicri:Area/Ncbi/Curation">000114</idno>
<idno type="wicri:Area/Ncbi/Checkpoint">000114</idno>
<idno type="wicri:doubleKey">1067-5027:2011:Rasmussen L:development:of:an</idno>
<idno type="wicri:Area/Main/Merge">000380</idno>
</publicationStmt>
<sourceDesc><biblStruct><analytic><title xml:lang="en" level="a" type="main">Development of an optical character recognition pipeline for handwritten form fields from an electronic health record</title>
<author><name sortKey="Rasmussen, Luke V" sort="Rasmussen, Luke V" uniqKey="Rasmussen L" first="Luke V" last="Rasmussen">Luke V. Rasmussen</name>
<affiliation wicri:level="2"><nlm:aff id="aff1">Biomedical Informatics Research Center, Marshfield Clinic Research Foundation, Marshfield, Wisconsin, USA</nlm:aff>
<country xml:lang="fr">États-Unis</country>
<wicri:regionArea>Biomedical Informatics Research Center, Marshfield Clinic Research Foundation, Marshfield, Wisconsin</wicri:regionArea>
<placeName><region type="state">Wisconsin</region>
</placeName>
</affiliation>
</author>
<author><name sortKey="Peissig, Peggy L" sort="Peissig, Peggy L" uniqKey="Peissig P" first="Peggy L" last="Peissig">Peggy L. Peissig</name>
<affiliation wicri:level="2"><nlm:aff id="aff1">Biomedical Informatics Research Center, Marshfield Clinic Research Foundation, Marshfield, Wisconsin, USA</nlm:aff>
<country xml:lang="fr">États-Unis</country>
<wicri:regionArea>Biomedical Informatics Research Center, Marshfield Clinic Research Foundation, Marshfield, Wisconsin</wicri:regionArea>
<placeName><region type="state">Wisconsin</region>
</placeName>
</affiliation>
</author>
<author><name sortKey="Mccarty, Catherine A" sort="Mccarty, Catherine A" uniqKey="Mccarty C" first="Catherine A" last="Mccarty">Catherine A. Mccarty</name>
<affiliation wicri:level="2"><nlm:aff id="aff2">Essentia Institute of Rural Health, Duluth, Minnesota, USA</nlm:aff>
<country xml:lang="fr">États-Unis</country>
<wicri:regionArea>Essentia Institute of Rural Health, Duluth, Minnesota</wicri:regionArea>
<placeName><region type="state">Minnesota</region>
</placeName>
</affiliation>
</author>
<author><name sortKey="Starren, Justin" sort="Starren, Justin" uniqKey="Starren J" first="Justin" last="Starren">Justin Starren</name>
<affiliation wicri:level="2"><nlm:aff id="aff1">Biomedical Informatics Research Center, Marshfield Clinic Research Foundation, Marshfield, Wisconsin, USA</nlm:aff>
<country xml:lang="fr">États-Unis</country>
<wicri:regionArea>Biomedical Informatics Research Center, Marshfield Clinic Research Foundation, Marshfield, Wisconsin</wicri:regionArea>
<placeName><region type="state">Wisconsin</region>
</placeName>
</affiliation>
<affiliation wicri:level="2"><nlm:aff id="aff3">Division of Health and Biomedical Informatics, Department of Preventive Medicine, Northwestern University, Chicago, Illinois, USA</nlm:aff>
<country xml:lang="fr">États-Unis</country>
<wicri:regionArea>Division of Health and Biomedical Informatics, Department of Preventive Medicine, Northwestern University, Chicago, Illinois</wicri:regionArea>
<placeName><region type="state">Illinois</region>
</placeName>
</affiliation>
</author>
</analytic>
<series><title level="j">Journal of the American Medical Informatics Association : JAMIA</title>
<idno type="ISSN">1067-5027</idno>
<idno type="eISSN">1527-974X</idno>
<imprint><date when="2011">2011</date>
</imprint>
</series>
</biblStruct>
</sourceDesc>
</fileDesc>
<profileDesc><textClass></textClass>
</profileDesc>
</teiHeader>
<front><div type="abstract" xml:lang="en"><sec><title>Background</title>
<p>Although the penetration of electronic health records is increasing rapidly, much of the historical medical record is only available in handwritten notes and forms, which require labor-intensive, human chart abstraction for some clinical research. The few previous studies on automated extraction of data from these handwritten notes have focused on monolithic, custom-developed recognition systems or third-party systems that require proprietary forms.</p>
</sec>
<sec><title>Methods</title>
<p>We present an optical character recognition processing pipeline, which leverages the capabilities of existing third-party optical character recognition engines, and provides the flexibility offered by a modular custom-developed system. The system was configured and run on a selected set of form fields extracted from a corpus of handwritten ophthalmology forms.</p>
</sec>
<sec><title>Observations</title>
<p>The processing pipeline allowed multiple configurations to be run, with the optimal configuration consisting of the Nuance and LEADTOOLS engines running in parallel with a positive predictive value of 94.6% and a sensitivity of 13.5%.</p>
</sec>
<sec><title>Discussion</title>
<p>While limitations exist, preliminary experience from this project yielded insights on the generalizability and applicability of integrating multiple, inexpensive general-purpose third-party optical character recognition engines in a modular pipeline.</p>
</sec>
</div>
</front>
</TEI>
</record>
Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Ticri/CIDE/explor/OcrV1/Data/Main/Merge
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 000380 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Main/Merge/biblio.hfd -nk 000380 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien |wiki= Ticri/CIDE |area= OcrV1 |flux= Main |étape= Merge |type= RBID |clé= PMC:3392858 |texte= Development of an optical character recognition pipeline for handwritten form fields from an electronic health record }}
Pour générer des pages wiki
HfdIndexSelect -h $EXPLOR_AREA/Data/Main/Merge/RBID.i -Sk "pubmed:21890871" \ | HfdSelect -Kh $EXPLOR_AREA/Data/Main/Merge/biblio.hfd \ | NlmPubMed2Wicri -a OcrV1
![]() | This area was generated with Dilib version V0.6.32. | ![]() |